Siliconsciousness: Who Will Find the First AI Jobs Bite the Most Painful?

with the new Cuba-Wan Capsule Machine from Chibu,

“with the best café from special occasions.”

Full-Mondic Arômen, Danck Innovative Pressbrute Technology, and with 17-Sorten Café for every match. Elé per Premium Café is already in the 19th century. And there is the Cuba-Capsule Machine in Diner Chibofiale and Chibode E. The very good?

"The Stoyer is very good." "That's a whole lot." "Cool."

"What does it taste like?"

"The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good."

"The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good."

"The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good." "The Stoyer is very good."

"The Stoyer is very good."

“To stay up to date on all the news that you need to know, there's no better place than right here on the DSR network.”

And there's no better way to enjoy the DSR network than by becoming a member. Members enjoying ad-free listening experience, access to our discord community, exclusive content, early episode access, and more. Use code in DSR 26 for 25% off discount on sign up at the DSR network.com. That's code in DSR 26 at the DSR network.com/buy. Thank you, and enjoy the show.

Welcome to "Silet Consciousness," the DSR network podcast focusing on the artificial intelligence revolution, politics, and policy. Hello, and welcome to "Silet Consciousness." I'm your host, David Rothkuff, and this week is every week. We'll be talking about an issue that's really important with regard to AI and related technologies. This week, we're extremely fortunate to have with us, George Eos Petropolis.

He's an assistant professor in the Department of Data Sciences and Operations of the USC Marshall School of Business.

“He's also a digital fellow at the MIT Initiative on the Digital Economy and the Digital Economy Lab at Stanford University.”

And a network affiliate at the CES, CES, IFO. Hi, George Eos. Thanks for joining us. Thanks so much for the invitation, looking forward to talking with you. You wrote an article for MIT Technology Review, our friends, and sometimes partners there.

And the article is entitled, "It's time to address the looming crisis in entry-level work." And while I'll leave it to you to flesh out the thoughts behind the article briefly, it argues that, while many people have been anticipating crises of different types with regard to dislocations associated with AI,

that the place it seems to be hitting first, or at least one place where there is some data that suggests that it's having an impact,

has to do with entry-level jobs in certain areas. Can you flesh this out a little bit? Yes, since the deployment and the mass-view show for large language models like Tajipiti, we have seen that some effects in the digital market. And these effects are negative, particularly for the young workers.

The ones that just join the Joe Market and they don't, they haven't built experience in the work they are specialized on. And the studies show, there is some agreement among the recent studies that there is a decline in the number of jobs and the employment of young people. And the fact that these decline arrives after the large language model and their introduction to our life, makes it very worrying about the impact of artificial intelligence on entry-level jobs,

of people that they are young, they just end their, the Joe Market and they haven't built the experience that need to be more proficient in the work they are doing, or they are focusing on specific tasks in their jobs, so which are very easily automated with generative AI tools.

If AI can do these tasks, then the company does not need you anymore, right?

So that is the issue. And we see that this is particularly the case in a specific occupation where generative artificial intelligence, the large language models are used a lot in software engineering, in computer science, in programming, in customer service. So there we see that the decline of employment or young employees is quite significant. So would you say this is a warning sign, a kind of canary in the coal mine, about broader effects across labor markets,

or would you say this is specific to the kinds of jobs that are most impacted or most easily replicable by large language. What is language models?

“I believe that the correct, the correct approach to see that is towards your second question is when a job is affected by AI by large language models,”

that means that large language models are used a lot in the job for producing some output. Then the young people, the ones that they are less experienced, are the ones that will see more risk of their job on their jobs, because exactly the large language models will be more capable to do the specific tasks they do in their job. I mean it's job, it's occupation, has a continuous series of tasks. There are the entry-level tasks, there are the tasks of more experienced people, more experienced worker.

So AI and large language models seems to replicate the entry-level tasks, and that makes it more challenging for young people to find a job in those occupations. But there is also an additional element here.

“It is also AI as a tool is remarkable tool, may people and especially young people more productive, right?”

So a firm that wanted in the past would hire two employees for delivering a given output. Nowadays, having one employee with the assistance of AI is more than enough to arrive to that output goal. So the fact that AI in large language model has been so influential and so good creates a problem, even if that sounds initially as paradoxical thing to happen. Well, yeah, I mean it creates a problem, but it also creates a problem with the current state of AI, right? So if AI develops further and develops additional capabilities, if AI learns more, perhaps it is not just the entry-level people who are targeting, right?

Certainly, although we can never be sure about the future, the large language models as a development was something that it was very hard to predict some years ago.

Even if you were into the industry and you were following the news, you were surprised by these new tools and how remarkable they are.

“It's hard to predict the future, but certainly, I believe we are in the beginning of a new transformation and you big transformation.”

And there are certainly more episodes that will follow and I believe that at some point it will not be only the entry-level that will be in question. The SAI becomes more capable and proceed further, at least in the occupations in the jobs that it is used, it can have a more significant role. That, of course, will generate some questions, for example, in the future AI may create also some other jobs for humans that we didn't know in the past.

We look at the technology and the history of technology always a technological breakthrough, replaces some jobs, replaces some tasks, but creates some other jobs and some other tasks to do.

I would say the overall saying is positive, so I'm not worried so much about the overall impact in employment. I believe that AI as a tool can generate also some new forms of employment, new forms of work, what I'm particularly worried is about the young people and whether they will have the opportunity to build some experience.

So there are two problems with that analysis. I'm not challenging the analysis. There are two problems that are associated with it. One is, you know, what do you do if you're a young worker today.

But now, according to the article, you know, we have seen a kind of softening of the entry level market ever since COVID. So this has made this a problem for several years and this is exacerbating that problem. And as you say, you know, in the past new technologies may have eliminated some jobs, but it created other jobs.

But of course, we don't know how quickly that's going to happen. And in the past that dislocation has not always been a comfortable one.

We're job, you know, if you worked in the fields because agriculture was not automated and then it was automated. And the fact that it was creating jobs in a city 500 miles away was small comfort. In other words, there are social and economic factors associated with these transitions that are not ameliorated by the possible future creation of jobs someplace else at an unspecific time.

“So I was wondering how how how young workers supposed to address that. Yeah, I think you're your question is very important.”

And I want beforehand to say that in its technological revolution, we always have winners and losers, right? So it's not that everyone will be better off.

There are challenges, it depends on what is your qualifications and what job you are, how it is affected by technology and AI, particularly. So the transformation is real and its transformation has some people that lose and some people that win the challenge as a society is to ensure that the people that win the winners are much more than the losers.

“Going on the challenge on the young workers, I believe the mentality of a young worker today has changed dramatically from the mentality five years ago.”

What I mean, young workers now they should from young kids, even during their education, during their education and the ERS, they spend in schools and universities to develop skills of collaboration with AI tools. That because especially in specific jobs that AI has a particular prominent role, having good understanding being able to collaborate in an efficient way with VCI tools, makes you to be the proper candidate for the future and company will prefer you to deal with you.

“I think having AI in curriculum, having been embedded how to build new skills based on collaborating with AI, it's a fundamental importance in order to be able to see a better future and to be more optimistic about the future.”

The second thing that is equally important is the ERS you spend in education should not be ERS that you are outside the job market and what do I mean by that practical experience experience in general matters. So that if you are in experience, you have harder time. So students should be motivated even from early ERS of their education in schools early university, in other grads, to look for this practical guidance, practical experience, which means also that the university should be more directed to this direction to provide such opportunity for the students.

So becoming more proficient on how to use AI tools in doing particular tasks and jobs and also finding opportunities for gaining some experience through some practice in companies and corporations while you are studying.

This podcast is underwritten and part by the US Embassy of the United Arab Emirates.

It's editorial content is completely independent and the views expressed are exclusively those of participating experts. It is presented live without editing. For further information about the UAE's efforts in the areas of artificial intelligence and technology, go to the website of the embassy at www.ue-emBC.org and search for UAE-US Tech Cooperation.

“We thank them for their support. We thank everybody who is supporting this podcast for their support and we look forward to it developing and growing over time because the issue is so important.”

What presumably a third approach would be to say what areas will be least affected by AI and try to go into more of an AI proof profession.

Yes, so there are professions that are affected less from AI and actually start to show that these jobs didn't find any significant drop in employment of young people.

“That's true. But I would say that the fundamental difference now with the technologies in the past is that AI is targeting more high-paid high-quality high-skill jobs.”

While in the past the technology was focusing on specific routine jobs, repetitive jobs, which are not the so-called high-skill jobs, for example construction workers or so-on.

We see that with the development of large language models, the cognitive ability of people with challenge, especially the ones that they are in experience. So yes, we can look for a job that is less affected by AI, but right now AI targets some jobs that traditionally were considered as high-paid high-skill jobs, so that means reduce the pool of available jobs in the high-skill high-wades domain if you would like to target a job that is not affected by AI. I mean, doctors is a great example. So a few years ago people were believing that radiologists will be replaced by AI. This never happened because the practical skills for providing this job task is really important, and this is by definition in human kind of job.

“So yes, directing to this kind of occupations can be a solution, but the pool of these occupations shrinks because AI affect high-paid and high-skill jobs.”

And you may also face greater competition if everyone thinks in the same way like you. Yeah, I'm a little skeptical, frankly, of statements that say this never happened when we're five years into a revolution, but because we don't know how it's going to evolve further and certainly one of the things that we've seen particularly in areas like radiology is that it's one of the areas in which being AI plus. In other words, a person who's able to use AI, you will get a better outcome, and that suggests to me a sort of changing nature of the way the service is delivered.

But let me ask you a question, you teach in that university, you teach in that business school, you also work in other university settings. You know what I think of what people learn at a business school, for example, they'll take courses in finance or they'll take courses in industrial organization or they'll take courses that involve certain kinds of research.

A lot of what they taught is actually, you know, replicable by AI. To what extent do you see educators responding to this?

And saying, oh, no, we have to teach a generation of AI plus workers, workers who know their field, but who know how to work with AI and can understand trends in AI and can anticipate those trends or or be among those who are creating new jobs using AI. It seems to me like as big a challenge as this is for the worker, it's an equally large challenge for the educated. Certainly. And in many educational institutions, there is an active discussion on what is the proper way forward.

“I believe that the, the role of the institution in the future will be fundamentally more important, not less important, but that will require some transformation that will fit better the needs of students in this year.”

And what I mean by that, so the first one, the metal shift we need to have and this is something ongoing is to provide the training in universities or students or how to use more efficient way, we say, I tools. The second is that we need to to build a practical, a practical part, an industry, a practical industry program in more and more university programs, so that students will be able to get a first hand experience on specific occupation that they are interested in.

“For example, you, you are absolutely right that in the case of radiologists, the, the use of AI by the radiologists created the created a much more efficient service, so there the value is in the interaction between humans and AI.”

Certainly, if the interaction brings this added value, you need to develop some experience on this. And I guess the third dimension which is also important is of course for young graduates, they, it will be welcome as they compete for fighting a job to belong to a network.

So building an expanded network of graduates and if school seriously invest also to this networking dimension, that can be helpful.

So more practical experience for students in the grading seriously AI tools on specific job tasks and trains students on how they use AI tools.

“And building a stronger basis for networking, I think that will be the free fall demandal aspects which can make education more important in the future and change helps students to find, survive this competitive pressure and find a good job.”

Yeah, of course, you know, in countries like the United States where education in particularly higher education is incredibly costly.

Our problems associated with this fewer and fewer people can afford what they need more and more. And furthermore, the models are very much oriented towards old models of employment. You know, you got trained as a doctor and then you did it for 40 years. But now, of course, if you have technologies like AI reinventing the fields on an ongoing basis, you can't get trained and then go and do it for 40 years. You need lifelong education. And if the government isn't providing that, if only those who are employed are getting that, it can be a gap creator in society rather than an opportunity creator in society.

One of the questions that I've got, we're running out of time here, but I'd like to ask you one more question. One of the questions that I've got is of course, what do you see is the public policy implications of these dislocations.

And, you know, you've seen similar results across the board because we don't know where the next generation of jobs will come because

“this does mean a social dislocation. In the past, what governments have done, as they said, well, we'll create a retraining program.”

Or we will create vocational education programs or other kinds of programs, we'll create a social safety net that enable people to get through that.

You live in Sweden and you lose your job because it's been faced out. There is a social program to take care of you. If you live in the United States, there is no program.

And what we've seen in the recent past, even in the course of the past 30 years, is that the drive towards both globalization and higher productivity has led to dislocations and we don't have. We don't have effective programs to help people deal with that, which is cause hardship.

And, and so I'm just wondering, as you look at this, if you were advising people who are developing public policy, where are they going to have to work?

“So, first of all, that is an important issue. It's very important and goes beyond this entry level of jobs that we discussed. I believe that I would frame it that we need a new social contract.”

This social contract should target people especially that lose from this transformation by having as a basis a safety net. That could be some basic unemployment salary for people that they were dislocated in a sudden way because of this transformation. But we should not stay there. The lifelong learning building, a training system, where it's a person, either a worker that wants to improve in the work, his performing or someone that lost the job and wants to find a new job that to be relevant in the job market of tomorrow should have access.

That is a responsibility of the state. That requires also some private initiative since corporations know much better how the skills evolve. And that could also be something that the educational institutions can be part of it since they have the infrastructure and the experience.

“What we need, I believe, for that, is to develop a clean framework, find a way to finance this framework because that is a major challenge.”

And then make sure, and that is a responsibility of the state, that will bring the educational institutions and the corporations as participants, so that we can make that work in the large scale. In some European countries and in some specific scales, we have seen some examples. The case of United States, the scale at stake is much bigger and therefore we need to have serious discussions towards this other new, because providing a new social contract will be necessary as you move forward. And of course, a new social contract requires functioning political system. We don't have one of those at the moment at the United States, it would be great to have one, but it also requires the resolution of certain kinds of inherent tensions.

So just to add in conclusion, we have Jen Z, young people entering the marketplace, and because of all the factors you've just said, those work, those young people are super AI skeptical. And at the same time, because, you know, if you leave it to sort of jungle capitalism, American capitalism, it stands right now, the dislocation is going to be extremely painfield, maybe for protracted periods of time, and maybe in ways that create even greater social and economic inequality. So there's attention there. The tension is, you know, do you want to slow down the embrace of AI, because it's going to cause this, but that may in fact make the country less competitive and may make individuals less competitive if they just say, I'm not going to be involved in that.

So this is an issue that's going to be with us a long time. I think the work that you've done here and described in this article, is therefore extremely relevant. I encourage people to go and to read the article, which you'll find at the MIT Technology Review, it's called it's time to address the looming crisis and entry level work.

“And of course, follow the work that Georgia is doing and others in this same field, Georgia is perhaps we can continue this discussion at some point in the future.”

I don't think it's going away anytime soon for now. Thank you very much. Thank you everybody for listening. Please join us again next week. Bye bye.

This was Siliconjustness, a production of the DSR network.

the government will not be able to pay for the fact that the government is in the right place.